What would happen if you put a vast amount of liquid metal into a custom cooling loop instead of water/coolant? What challenges would you face? Would there even be any benefit to doing this?
BONUS: What if you used copper tubes instead of standard plastic/glass tubes and pumped liquid metal thro...
@Máté Juhász I cannot see why this was down-voted or put as on hold. This is clearly ON TOPIC for Superuser. It's also a good question with a clear and concise answer. Maybe not the best question in terms of practicality, but a perfectly valid question nonetheless... This IS about "computer hardware" as it has to do with water-cooling loops and components and how they'd be affected in this situation. If it's "off-topic," move it to the correct stack site..
I just upvoted. This is an interesting question but the source of where this question came from is the issue in my mind. Was it just a thought? Or you think liquid metal would look cool?
@JakeGould just a thought, I didn't consider what it might do to the tubes in the loop or anything, just a curious thought. Liquid metal is used for heat transfer from the CPU die to the IHS and is supposedly really good at it. So I thought why not in a CPU loop? In retrospect it seems obvious, but at the time I didn't think there'd be a huge difference in issues between using water & liquid metal. Oh boy I was wrong.
Something like this is used in nuclear reactors to cool them: youtube.com/watch?v=9EGAXOWpGy8 but NaK is some scary stuff and definitly nothing for home-usage :)
I wonder if we can get physics SE to help answer this question... I mean sure the theme of the question is for PC cooling but it is more experimental physics than PC knowledge required to answer this...
"If it's 'off-topic,' move it to the correct stack site" Assuming there is one! Not everything has to fit somewhere on Stack Exchange (not "stack"). However I'd agree this question is on-topic here.
@Pimp Juice IT: Not surface tension, it's because mercury is actually denser than most common metals: sites.google.com/site/chempendix/densities-of-pure-metals Indeed, you could float lithium, sodium or potassium metal on water, if you could keep them from reacting.
@TobiasMädel: NaK is a salt (an ionic compound), not a metal. Yes potassium is a metal, but the chemical bonds don't break when it melts.
@DavidPostill Obviously you'd have to choose a metal with a low enough melting point to be liquid at around room temperature, otherwise you'd be heating your CPU with liquid metal. It doesn't make sense to consider a metal that would have to be tube-meltingly hot to be liquid. (Unless you meant "melt" in the sense of a chemical reaction that softened and "melted" the plastic, but I don't think that would happen.)
This is really a design question; it'd be best on SE.Engineering. The stuff about moving heat around and designing coolers like this is well within the technical know-how of college-educated engineers. Specifically, it's studied in courses like "Transport Phenomena", "Fluid Dynamics", and "Thermodynamics". I'd note that engineers will tend to be familiar with this sort of system; physicists and chemists also have some conceptual background on it, but their education and research tend to focus on other topics.
The funny part is that, if with a little explaining, we could mostly convert this problem into an Electrical Engineering analog. Temperature's kinda like voltage; heat flow's kinda like current; the cooling loop is kinda like a resistor; the film between the liquid bulk phase and cooling block's solid phase are kinda like a junction resistance; the CPU and heat block themselves are like series resistors; etc.. This really would be a more fun problem to tackle in a site with TeX enabled, but it's so hard to type anything technical in plain text.
Discussion on question by FatalSleep:…
Discussion on question by FatalSleep:…